nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 23 Packages Projects Releases Wiki Activity

replace the magic nunber 768 by max work group size to support iGPU (#19920)

Browse Source 
Co-authored-by: Neo Zhang Jianyu <jianyu.zhang@intel.com>
This commit is contained in:
Neo Zhang
2026-02-27 09:26:07 +08:00
committed by GitHub
co-authored by Neo Zhang Jianyu
parent 88cf781f51
commit c17dce4f5c
1 changed files with 5 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-sycl/add-id.cpp
+5 -1
View File
@@ -55,7 +55,11 @@ void ggml_sycl_add_id(ggml_backend_sycl_context& ctx, ggml_tensor* dst) {
const int32_t* src2_d = (const int32_t*)src2->data; const int32_t* src2_d = (const int32_t*)src2->data;
float* dst_d = (float*)dst->data; float* dst_d = (float*)dst->data;
int threads = std::min((int)ne00, 768); // cols const unsigned int max_work_group_size = ggml_sycl_info().max_work_group_sizes[ctx.device];
assert(work_group_size % (WARP_SIZE * WARP_SIZE) == 0);
int threads = std::min((unsigned int)ne00, max_work_group_size); // cols
ctx.stream()->parallel_for( ctx.stream()->parallel_for(
sycl::nd_range<3>( sycl::nd_range<3>(
sycl::range<3>(1, ne02, ne01) * sycl::range<3>(1, 1, threads), sycl::range<3>(1, ne02, ne01) * sycl::range<3>(1, 1, threads),